WO2013119991A1

WO2013119991A1 - Interference suppression apparatus and method for femtocell

Info

Publication number: WO2013119991A1
Application number: PCT/US2013/025394
Authority: WO
Inventors: Jung-Tao Liu
Original assignee: Ablaze Wireless Corporation
Priority date: 2012-02-10
Filing date: 2013-02-08
Publication date: 2013-08-15

Abstract

ABSTRACT The invention discloses an interference suppression apparatus and method for femtocell. It can address the random problem by adding some smartness into the femtocell/small cell devices to narrow the transmit power pattern and hence reduce unwanted radiation and its effect to the existing network elements and address the random on/off nature by adjusting certain transmit characteristic of these small basestations in the goal to optimize some chosen network performance metrics. In addition, it can also alleviate the above-mentioned uplink interference by designing the basestation to be aware of such interference and actively reject it or passively avoid it and improve the robustness of the TFCI decoder by reducing its block error rate (BLER).

Description

INTERFERENCE SUPPRESSION APPARATUS AND METHOD FOR FEMTOCELL

TECHNICAL FIELD

The invention relates to femtocell (a.k.a. small cell); in particular, to an interference suppression apparatus and method for femtocell. However, for skilled artisan, the invention can be easily applied to any type of cellular basestations.

BACKGROUND ART

Recent development of femtocell (or sometimes called small cell) technologies has added a new challenge to the existing cellular network infrastructures. The femtocell has some unique characteristics such as users can randomly turn it on and off, and move it around as they wish to any location allowed by the Operators. These characteristics cause quite a challenge as these small devices often transmit at the same frequency as the existing large scale cellular network and sometimes cause unwanted interference to it.

Other newly introduced challenges include a mobile station near by the small cell (or femtocell) transmitting at high power only not to communicate with the small cell but with the far away cell tower. This high power signal could practically drown out the other mobile stations talking to the small cell causing uplink receiving problem at the small cell.

The signal received by the basestation is from multiple transmitting sources (mobile stations). This includes wanted signals as well as unwanted signals. The wanted signals and the unwanted signals are separately dealt with. One of the signal processing techniques that is commonly used to suppress interference and perform joint detection involves calculating the correlation matrix of the signals received and use that in either linear or non-linear joint detection algorithms like MMSE detector and ML or MAP detectors. However, this is not usually feasible if the mobile stations are transmitting with signals that come as burst data since it usually takes some time for the correlation matrix estimation to converge. In UMTS (universal mobile terrestrial system) or 3G, a data channel must accompany a control channel which carries information about how the data channel is manipulated and carried over the air. One of such control information is called the TFCI (Transport Format Combination Indicator). To be able to process the received data, a receiver must first decode the TFCI carried on the control channel correctly.

In addition, network scanning/monitoring is a unique requirement for femtocell. Due to the ad hoc nature of its deployment scenario, conventional network planning which is usually done manually now requires automation. Such requirements including the establishment of neighboring cell information, location information, intra-cell interference optimization, etc. Network scanning should be performed over other interfering systems such as GSM, WCDMA, LTE, and GPRS. The TR-196 specification does not specify how often such tasks should be perfumed but does specify a time out mechanism which limits the amount of time that can be used to perform the scanning and monitoring.

Therefore, the invention provides an interference suppression apparatus and method for femtocell to solve the above-mentioned problems occurred in the prior arts.

DISCLOSURE OF INVENTION

A scope of the invention is to address the above-mentioned random on/off problem by adding some smartness into the femtocell and small cell devices to narrow the transmit power, and signal or antenna pattern and hence reduce unwanted radiation and its effect to the existing network elements. Different methods are used to adjust the transmit signal pattern from the femtocell for channels that are common to all mobiles communicating with the femtocell (we call these the common channels) and for channels or transmissions that are dedicated to a particular mobile (we call these the dedicated channels).

Another scope of the invention is to address the above-mentioned random on/off nature by adjusting certain transmit characteristic of these small basestations (in the following figure, we call them basestation X) in the goal to optimize some chosen network performance metrics. The invention discloses multiple cellular basestations capable of communicating with a optimization units which calculate various parameters for each of the basestations under control and configure the basestation to transmit its signal at a certain power level at a certainly frequency and in a certain direction using certain signal pattern in a continuous faction aiming to increase overall system performance in turns of, for example, enhancing capacity, throughput, or lowering interference to other basestations. The innovative part of this invention is that, there are cellular basestations that are not communicating to the optimization module and hence cannot be adjusted, the optimization module nonetheless, optimizes the cellular network performance by taking into account of all information available about the environment, such information are collected by basestations it is communicating with regarding all basestations, and adjusting the basestations that is controllable by the optimization unit while leaving the original/incumbent basestations intact.

Another scope of the invention is to alleviate the above-mentioned uplink interference by designing the basestation to be aware of such interference and actively reject it or passively avoid it.

Another scope of the invention is to improve the robustness of the TFCI decoder by reducing its block error rate (BLER).

In an embodiment, an interference suppression apparatus for a femtocell is disclosed. The femtocell has a minimum of two received antenna and RF paths and a minimum of two transmit antenna and RF paths. The interference suppression apparatus includes a monitoring device for monitoring interfering signals from different directions. The monitoring device includes a forming module, a selecting module, and a RSSI calculating module. The forming module is used for forming a set of complex number pairs. The selecting module is coupled to the forming module and used for selecting a first complex number pair from the set of complex number pairs. The RSSI calculating module is coupled to the selecting module and used for calculating a first RSSI value corresponding to the first complex number pair. Then, the selecting module selects a second complex number pair next to the first complex number pair from the set of complex number pairs, and the RSSI calculating module calculates a second RSSI value corresponding to the second complex number pair.

In an embodiment, an interference suppression method for a femtocell to monitor interfering signals from different directions is disclosed. The femtocell has a minimum of two received antenna and RF paths and a minimum of two transmit antenna and RF paths. The interference suppression method includes steps of: forming a set of complex number pairs; selecting a first complex number pair from the set of complex number pairs; calculating a first RSSI value corresponding to the first complex number pair; selecting a second complex number pair next to the first complex number pair from the set of complex number pairs; calculating a second RSSI value corresponding to the second complex number pair.

In an embodiment, an interference suppression apparatus for a femtocell is disclosed. The femtocell can be arbitrarily powered on or off. The interference suppression apparatus includes a gateway/aggregator, a configuration server, and an optimization module. The femtocell is connected to the gateway/aggregator. The configuration server may be coupled to the gateway/aggregator and is communicating with the femtocell. The optimization module may be coupled to the configuration server and is used for receiving information related to the femtocell from the configuration server or directly from the femtocell and continuously adjusting parameters of the femtocell according to the information related to the femtocell to optimize system performance metrics.

In an embodiment, an interference suppression method for a femtocell is disclosed. The femtocell can be arbitrarily powered on or off. The interference suppression method includes steps of: using a protocol to communicate with the femtocell; receiving information related to the femtocell; receiving information related to the femtocell from the configuration server; continuously adjusting parameters of the femtocell according to the information related to the femtocell to optimize system performance metrics.

In an embodiment, an interference suppression apparatus for a femtocell is disclosed. The interference suppression apparatus includes at least one receiving antenna, at least one RF receiver, at least one multi-channel channel estimation module, at least one fast Fourier transform (FFT) module, a correlation matrix calculator, a correlation matrix estimator, and a joint detection module. The at least one RF receiver is coupled to the at least one receiving antenna. The at least one multi-channel channel estimation module is coupled to the at least one RF receiver and used for receiving a signal from the at least one RF receiver and forming an augmented channel matrix. The at least one fast Fourier transform (FFT) module is coupled to the at least one multi-channel channel estimation module and used for receiving output of the at least one multi-channel channel estimation module to form a transform domain signal. The correlation matrix calculator is coupled to the at least one FFT module and used for generating a calculated correlation. The correlation matrix estimator is coupled to the correlation matrix calculator and the at least one RF receiver and used for receiving output of the at least one RF receiver to estimate a received signal correlation and converting the received signal correlation to transform domain. The correlation matrix estimator receives the calculated correlation from the correlation matrix calculator and subtracts the calculated correlation from an estimated correlation matrix. The joint detection module is coupled to the at least one RF receiver, the at least one FFT module, the correlation matrix calculator, and the correlation matrix estimator and used for jointly detecting user signals.

In an embodiment, a transport format combination indicator (TFCI) decoder used in universal mobile terrestrial system (UMTS) is disclosed. The TFCI decoder includes a mask generator, a TFCI decoder engine, a decision device, a TFCI value calculator, a TFCI validation module, and a transmission time interval (TTI) level TFCI decoder. The mask generator is used for generating masks. The TFCI decoder engine is coupled to the mask generator and a symbol demodulator and used for receiving the masks and an input from the symbol demodulator to generate a vector of outputs. The decision device is coupled to the TFCI decoder engine and used for searching a maximum value of the outputs and recording the maximum value, index points to the position of the maximum value in the vector, and a mask ID. The TFCI value calculator is coupled to the decision device and for receiving the maximum value and reconstructing a TFCI signal. The TFCI validation module is coupled to the TFCI value calculator and used for validate whether the reconstructed TFCI signal has a legal value, if yes, the TFCI validation module setting a valid fail flag to false. The transmission time interval (TTI) level TFCI decoder is coupled to the TFCI validation module and used for collecting TFCI values from the TFCI validation module belong to the same transmission time interval and outputting a specific TFCI value which gets the highest vote counts among the TFCI values.

In an embodiment, an interference suppression method for a femtocell to continuously monitor in-band interference to optimize a transmit pattern of the femtocell is disclosed. The interference suppression method includes steps of: shutting down a RF transmitter of the femtocell when a measurement is taken to avoid a RF front-end saturation; if the interference suppression method is operated in a reset mode, normal femtocell transmission and reception being interrupted, a PHY chip itself being reset, all counters and internal states are wiped clean to initiate sniffing, software protocol being not interrupted, after pre-defined sniffing period has passed, the normal femtocell operation will be restarted; if the interference suppression method is operated in a stealth mode, normal femtocell operation being muted, all downlink and uplink transmissions/receptions being blanked out, the interference suppression method creating a transmission gap affecting the downlink transmission.

Compared to the prior arts, the invention can address the above-mentioned random on/off problem by adding some smartness into the femtocell/small cell devices to narrow the transmit power pattern and hence reduce unwanted radiation and its effect to the existing network elements and address the above-mentioned random on/off nature by adjusting certain transmit characteristic of these small basestations (in the following figure, we call them basestation X) in the goal to optimize some chosen network performance metrics. In addition, the invention can also alleviate the above-mentioned uplink interference by designing the basestation to be aware of such interference and actively reject it or passively avoid it and improve the robustness of the TFCI decoder by reducing its block error rate (BLER).

The advantage and spirit of the invention may be understood by the following detailed descriptions together with the appended drawings.

BRIEF DESCRIPTION OF DRAWINGS

FIG. 1 shows a possible implementation of a monitoring device that can listen to interfering signals from different directions.

FIG. 2 shows a possible implementation of the common channel signal pattern adjustment for a 3G UMTS cellular basestation.

FIG. 3 shows a possible implementation of phase and amplitude adjustment for dedicated channel.

FIG. 4 shows a Markov chain state transition diagram

FIG. 5 shows a weight generator implementing the Markov chain.

FIG. 6 shows a diagram of an exemplary implementation of the entities involved.

FIG. 7 shows a diagram of an exemplary implementation of an interference suppression apparatus.

FIG. 8 shows a diagram of an exemplary implementation of the TFCI decoder.

FIG. 9 shows the minimum length of GP1.

FIG. 10 shows an overview of the windows for GSM sniffer operations.

FIG. 11 shows the IPDL RF characteristics. FIG. 12A-12B show the flowchart of an example of complete GSM continuous sniffing operation method.

FIG. 13 shows the minimum operating window for WCDMA in-band sniffing.

FIG. 14A-14C show the flowchart of an example of complete WCDMA continuous sniffing operation method.

BEST MODE FOR CARRYING OUT THE INVENTION

(I) Downlink Interference Mitigation using dynamic subspace tracking and beamforming for cellular network with small basestation with random on/off characteristics

In order to address the above-mentioned random on/off problem, some smartness is added into the femtocell/small cell devices to narrow the transmit power pattern and hence reduce unwanted radiation and its effect to the existing network elements. Different methods are used to adjust the transmit signal pattern from the femtocell for channels that are common to all mobiles communicating with the femtocell (we call these the common channels) and for channels or transmissions that are dedicated to a particular mobile (we call these the dedicated channels)

In this embodiment, a basestation having the capability of monitoring interfering signal from different directions is provided. And, the basestation has a minimum of two received antenna and RF paths and a minimum of two transmit antenna and RF paths. Please refer to FIG. 1. FIG. 1 shows a possible implementation of a monitoring device that can listen to interfering signals from different directions.

As shown in FIG. 1, the monitoring device MD operates as follows:

(l)Forming a pre-defined set of complex number pairs (Z1,Z2);

(2)Selecting the first pair inside the set of (Z1,Z2) pairs and calculating a first RSSI value; (3) Moving on to the second pair inside the set of (Z1,Z2) pairs and calculating a second RSSI value;

(4) Continually doing so until the end of the set of (Z1,Z2) pairs.

Then, please refer to FIG. 2. FIG. 2 shows a possible implementation of the common channel signal pattern adjustment for a 3G UMTS cellular basestation. As shown in FIG. 2, the weights Wl and W2 are determined as follows:

(1) From the set of (Zl, Z2) pairs, the basestation selects a pair of (Zl, Z2) with minimum RSSI value calculated by the monitoring device MD shown in FIG. 1 ;

(2) Set (W1,W2) according to the pair of (Zl, Z2).

Please refer to FIG. 3. FIG. 3 shows a possible implementation of phase and amplitude adjustment for dedicated channel. In this invention, two possible methods of adjusting dedicated channel pattern are provided. Please refer to FIG. 4 and FIG. 5. FIG. 4 shows a Markov chain state transition diagram; FIG. 5 shows a weight generator implementing the Markov chain. As shown in FIG. 4, in the Markov chain, (Wl, W2) pairs are adjusted gradually from one coordinate to another.

In the first method, it operates as follows:

(1) As shown in FIG.5, the weight generator WG receives binary control bit sequence from the basestation and generates a complex valued (Wl , W2) pair;

(2) Pre-defined possible values for (W1,W2) pairs, labeled as 1,2,3, and 4 are shown in FIG. 4. These labeled values are sometimes also called the states of a Markov process;

(3) When the basestation starts to adjust the dedicated channel transmit pattern, it starts with state 1, slot number 0 in the transmission frame;

(4) The basestation keep the (W1,W2) pair at state 1 by input appropriate binary bit pattern to the weight generator WG;

(5)Basestation getting quality indicator feedback from the mobile station it has the dedicated connection with, such as TPC command sequences, Channel quality indicator, etc, in a 3G UMTS mobile system; (6) Basestation use the binary control bit sequence to traverse all possible states in the Markov process and record the quality indicator feedback from the mobile station having the dedicate connection with;

(7) After scanning through all states, the basestation selects a state that provides the best quality indicator;

(8) Basestation periodically repeat the scanning process defined in steps (5), (6), and (7) to ensure the (W1,W2) pair selected generates the best quality indicator feedback from the mobile station having the dedicate connection with the basestation.

In the second method, it operates as follows:

(l)With a set of pre-defined set of complex valued pair (Wl , W2);

(2) The basestation selects the first pair and records the quality indicator feedback from the mobile it has the dedicated connection with;

(3) The basestation selects the next pair and repeat step (2), until it runs through the whole set;

(4)The basestation selects the best pair with the best quality indicator feedback from the mobile it has the dedicated connection with;

(5)The basestation periodically repeats steps (3) and (4) to ensure the (W1,W2) pair selected generates the best quality indicator feedback from the mobile station it has the dedicated connection with.

(ID Downlink interference control and optimization in hierarchical cellular network

In this embodiment, there are cellular basestations that are not communicating the optimization module and hence cannot be adjusted, the optimization module nonetheless, optimizes the cellular network performance by taking into account of all information available about the environment, such information are collected by basestations it is communicating with regarding all basestations, and adjusting the basestations that is controllable by the optimization unit while leaving the original/incumbent basestations intact. Please refer to FIG. 6. FIG. 6 shows a diagram of an exemplary implementation of the entities involved. As shown in FIG. 6, basestations X are a group of basestations that can be configured by the optimization module OPM, and the basestations X are possibly connected to a gateway/aggregator GA first before connecting to the optimization module OPM. Basestations Y are incumbent infrastructures possibly connecting to a radio network controller RNC. A configuration server CS is able to communicate with basestations X using protocols such as TR069 or TR196. Optimization module OPM can receive information from the configuration server CS.

One example of such information including but not limited to reporting basestation's Neighboring basestation ID, reporting basestation's Neighboring basestation RSSI (received signal strength indicator), reporting basestation's connected mobile station's quality indicator such as block error rate (BLER), reporting basestation's power control command for a connected mobile station in the certain period of time from the past, or reporting basestation's Neighboring basestation current antenna pattern angles, azimuth.

The basestations X can be arbitrarily power on or off. Optimization module OPM can receive constant information updated from the basestations X through either configuration server CS. The configuration server CS can either be a part of the gateway/aggregator GA or as a separate entity/hardware. The optimization module OPM can continuously adjust various parameters of the basestations X in order to optimize overall system performance.

The optimization module OPM can decide at least the following information for a given basestation X: antenna phase multiplier, azimuth multiplier, angle multiplier, and transmission power. The optimization module OPM can be designed to optimize different performance metrics, such as system throughput, perceived interference, user quality indicator like CQI, RSSI reported by mobile station, but not limited to these.

(HI) Uplink interference suppression and rejection for wireless system A mobile station near by a small cell (or femtocell) transmitting at high power not only to communicate with the small cell but also with the far away cell tower. This high power signal could practically drown out the mobile stations talking to the small cell causing uplink receiving problem at the small cell. In this embodiment, this uplink interference is suppressed by designing the basestation to be aware of such interference and actively reject it or passively avoid it.

Please refer to FIG. 7. FIG. 7 shows a diagram of an exemplary implementation of an interference suppression apparatus. As shown in FIG. 7, the interference suppression apparatus 7 includes at least one receive antenna 70, at least one RF receiver 71, a correlation matrix estimator 72, a correlation matrix calculator 73, at least one multichannel channel estimation module 74, at least one fast Fourier transformation (FFT) module (or other kinds of orthogonal transformation module) 75, and a joint detection module 76. Wherein, the receive antenna 70 is coupled to the RF receiver 71; the RF receiver 71 is coupled to the correlation matrix estimator 72, the multi-channel channel estimation module 74, and the joint detection module 76; the correlation matrix estimator 72 is coupled to the correlation matrix calculator 73 and the joint detection module 76; the correlation matrix calculator 73 is coupled to the FFT module 75 and the joint detection module 76; the multi-channel channel estimation module 74 is coupled to the FFT module 75; the FFT module 75 is coupled to the joint detection module 76.

A multi-channel channel estimation module 74 receives signals from the at least one

RF receiver 71 and forms an augmented channel matrix and estimates the channel based on pilots of known transmit symbols. The outputs of the multi-channel channel estimation modules 74 are fed to the FFT modules 75 to form transform domain signals which are further processed by the correlation matrix calculator 73. The correlation matrix estimator 72 receives signals from the at least one RF receiver 71 to estimate the receive signal correlation and then convert it to the transform domain. The correlation matrix estimator 72 receives the calculated correlation from the correlation matrix calculator 73. The calculated correlation is subtracted from the estimated correlation matrix and then fed into a bank of IIR (infinite impulse response) filters. The outputs of the filter bank are passed to the joint detection module 76. The joint detection module 76 receives input signals from the RF receiver 71, the FFT module 75, the correlation matrix estimator 72, and the correlation matrix calculator 73. The inputs from the correlation matrix estimator 72 and the correlation matrix calculator 73 are added together to form a matrix A. Matrix A and the signals from the multi-channel channel estimation module 74 and the RF receiver 71 are further processed to jointly detect the user signal. The signal received by the RF receiver 71 is first multiply by the complex conjugate of the multi-channel channel estimation module 74 and then divided by matrix A to form a signal B and then the signal B is transmitted to an IFFT module (inverse FFT). The steps of forming the signal B and transmitting the signal B to the IFFT module are repeated multiple times until all mobile station signals are detected.

(IV) Improved UMTS TFCI decoder

In this embodiment, the block error rate (BLER) of the TFCI (Transport Format Combination Indicator) decoder will be reduced to improve the robustness of the TFCI decoder. Please refer to FIG. 8. FIG. 8 shows a diagram of an exemplary implementation of the TFCI decoder. As shown in FIG. 8, the TFCI decoder 8 includes a mask generator 80, a TFCI decoder engine 81, a decision device 82, a TFCI value calculator 83, a TFCI validation module 84, and a TTI (transmission time interval) level TFCI decoder 85. Wherein, the mask generator 80 is coupled to the TFCI decoder engine 81; the TFCI decoder engine 81 is coupled to the decision device 82; the decision device 82 is coupled to the TFCI value calculator 83 and the TFCI validation module 84; the TFCI value calculator 83 is coupled to the TFCI validation module 84; the TFCI validation module 84 is coupled to the TTI level TFCI decoder 85. In fact, the TFCI decoder engine 81 can be an inverse fast hadamard transforming decoder engine, but not limited to this. The input from the symbol demodulator is transmitted to the TFCI decoder engine 81. Within the TFCI decoder engine 81 , the mask generator 80 generates all possible masks. TFCI decoder engine 81 uses these masks to generate a vector of outputs. The decision device 82 searches for the maximum within these outputs, records its maximum value, the index points to the position of the maximum value in the output vector, and the mask id. The TFCI value calculator 83 receives these values and reconstructs the TFCI. Then, this TFCI is transmitted to the TFCI validation module 84 to validate whether the reconstructed TFCI has a legal value.

If the validation result is Yes and the TFCI decoder engine 81 never failed before, the TFCI validation module 84 sets the valid fail flag to false and sends the TFCI value to the TTI level TFCI decoder 85.

If the validation result is No, the TFCI validation module 84 sets the valid fail flag to true. If this is the first time failed the validation, the TFCI validation module 84 saves the maximum value, the index points to the position of the maximum value in the output vector, and mask id found. And, the TFCI validation module 84 sets element inside the output vector corresponding to the maximum value found in 4 to 0, and searches for the maximum value again, then outputs re-record its maximum value, the index points to the position of the maximum value in the output vector, and mask id, and The TFCI value calculator 83 receives these values and reconstructs the TFCI.

If all possible elements of the output vector from 4 are exhausted, restoring the maximum value saved. The TFCI validation module 84 sends the TFCI value to the TTI level TFCI decoder 85. The TTI level TFCI decoder 85 collects multiple TFCI outputs belong to the same TTI. And, at the TTI boundaries, it performs the following steps of: (1) processing all TTIs using a majority rules; (2) each TFCI's vote count is increased by X if a TFCI with that value is received, where X = 1, if valid fail flag = false; X < 1, if valid fail flag = true; (3) selecting the TFCI values getting the highest vote count as the output; if there is a tie between different TFCI values, randomly selecting a TFCI value from those having the same vote count as the output.

(V) Method for Continuous In-band GSM and WCDMA Interference Monitoring in Wireless Device

Continuous 2G and 3G network monitoring is a desired feature for femtocell, unlike the traditional cellular network, femtocell deployment is perceived to be in a random and ad-hoc fashion where end users can activate and deactivate femtocell, move it around freely at any time of the day. By monitoring the radio environment continuously, femtocell can try to optimize its transmit pattern to avoid interference with neighboring femtocells or macro basestations.

When monitoring inter-frequency channels, continuous operations depend less on the femtocell transmit frequency. This is because the femtocell inter-frequency interference albeit can still be huge, can be filtered by a well-designed adjacent channel suppression filter. For intra-frequency sniffing, femtocell transmitter must be muted in order to avoid saturating and interfering with the sniffer receiver operation.

Sniffer operations (both GSM and WCDMA) can be categorized into inter- frequency and intra-frequency sniffing. Inter-frequency monitoring are somewhat easier as the transmitter used by normal femtocell operation and the receiver used to perform scanning are tuned to the different frequency bands and the possibility of causing interference is much less than for the case of intra-frequency monitoring. This means that the inter- frequency scanning can be performed without interfering with the normal femtocell operation.

Intra-frequency (or in-band) monitoring requires that the transmitter RF be shut down while the measurement is being taken to avoid RF front-end saturation. Sniffing is envisioned for both power on initial scan and periodic scanning during normal femtocell operation. Depending on whether there is a dedicated received RF chain for sniffer, the operation can then divided into reset mode and stealth mode. During the reset mode, the normal femtocell transmission and reception is interrupted, PHY chip itself is reset, all counters, and internal states are wiped clean to initiate sniffing. Software protocol, however, is not interrupted. After pre-defined sniffing period has passed, the normal femtocell operation shall be restarted.

For stealth mode, the normal femtocell operation is "muted", using a mechanism similar to IPDL but with much longer idle period. During the stealth mode, all DL and UL transmission/reception are blanked out. The stealth mode creates a transmission gap which affects the downlink transmission. If there is a UE in CELL DCH mode, it is not recommended to have the gap greater than 10 radio frames (100 ms). If longer gap is needed to complete the sniffing operation, it is recommended to create this gap when there is no UE in CELL FACH or CELL DCH modes. UE in CELL_PCH mode is OK as the FW can create the gap in between the PCH signals.

Also, network sniffing does not always need to be completed with a complete BCH decode. Sometimes it is equally important to simply do a quick scan to determine whether there is a new cell props up nearby. In such case, the stealth mode is always preferred especially for intra-frequency sniffing.

For inter-frequency sniffing, since it doesn't affect the normal femtocell operation, stealth mode is always preferred except possibly during the power on initial scan. The goal of the sniffing operation is at its minimum to obtain the neighboring cell's IDs, and other system related information. This often means the sniffer must decode the broadcast channel from the interfering cells.

In GSM, there is no IPDL; a proprietary radio "off pattern is designed for continuous sniffing while not affecting the femtocell operations. This off pattern cannot be too short as to not able to achieve the goal of decoding the BCCH from GSM. The GSM minimum received data collection window is defined as follows. Frequency Burst (FB) in GSM will happen every 10 frames except towards the end of the 51 -multi-frame where the FBs are separated by 11 frames. Taking into account the followings:

(l)the filter settling time which is approximately 300 samples in our design;

(2)RF transmitter ramp up and down time (depending on the RF chip specifications;

(3)one optional extra frame after FB frame for SCH detection;

In order to perform FB and SB detection, a minimum of 4.615*12+300/4*0.00369 = 55.65675 msec of data is needed. For FB only, the minimum is 4.615* 11+300/4*0.00369 = 51.04 msec. Note that if FB is detected near the end of the data windows, one may have to store the rough timing boundary and wait for the next SB for SB detection unless optional extra frame in item above is included in GP1. The minimum length of GP1 is illustrated in FIG. 9.

FIG. 10 shows an overview of the windows for GSM sniffer operations. Each gap means data samples are needed for GSM sniffer operations. This translates to transmission gap needed for intra-frequency GSM sniffing. GP1 (Gap pattern 1) must be > 55.65675 msec. GP2 must be greater than 577 usee (one GSM slot), is mandatory only if SB detection isn't possible in GP1. GP3 is formed by two groups of 4 gaps each of the gaps is at least one slot duration (577 usee). GP4 is similar to GP3 and is used if SI 13 is present.

Note that, all gaps must take into account the RF ramp up/down periods which are available from RF vendors. The biggest impact to support the continuous intra-frequency monitoring to the normal femtocell operation is caused by GP1, during which, approximately 5*15+8.4 = 83.4 WCDMA slots (5.566 radio frames) will be silenced. It would be interesting to know if PCCPCH has a period that can be used for GP1 with less impact to the overall system performance. For DPCH, the impact probably cannot be avoided, and UE might notice it during the voice call. The FB detector requires 300 samples of settling time before generating meaningful outputs. The total FB detection span is at least 51msec as suggested in the prior section. To ease this requirement, one can break up the 51 msec into a series of smaller time units. We can break up GP1 into 88 (or 96 if SB detection must be guaranteed) discontinuous miniature gaps each is approximately a GSM slot duration, spread them over a longer period. According to TS25.104, the IPDL RF ramp up and down time needs to follow the IPDL RF characteristics shown in FIG. 11.

FIG. 12A and FIG. 12B shows the flowchart of an example of complete GSM continuous sniffing operation method. As shown in FIG. 12, the method performs steps S 10 and S 11 to receive the GSM SNIFFER REQ (GSM sniffer request) message from the caller and the receiver starts reading the ARFCN (Absolute Radio Frequency Channel Number) that it is supposed to scan. In step SI 2, the method determines whether the receiver is already powered on or not, it selects a "power on" mode or "normal mode"; the receiver will decide whether there is a need to create transmission gap. For "normal mode", the method performs step SI 3, the receiver selects either In-band or Out-of-band mode. For "power on" mode, the method performs step S14 to set CREATE GAP = 0. If in-band mode, the method performs step S16 to set CREATE GAP = 1. If Out-of-band mode, the method performs step SI 5 to determine whether the second RF is available for scanning. If the second RF is available for scanning, the method performs step S14 to set CREATE_GAP = 0. If the second RF is not available for scanning, the method performs step S16 to set CREATE GAP = 1.

Then, in step SI 7, the method judges whether CREATE GAP = 1. If the judging result of the step SI 7 is yes, the method performs step SI 8 to create a first gap GP1. And then, the method performs steps SI 9 and S20 to turn on frequency burst (FB) detection and judge whether FB is detected. If No, back to the step S 17; if yes, the method performs step S21 to obtain the ARFCN Band Indicator of the FB detected and store it and step S22 to turn on FO compensation. Afterward, the method performs step S23 to judge whether CREATE GAP =1. If the judging result of the step S23 is yes, the method performs step S24 to judge whether the synchronization burst (SB) detection period is included in the first gap GPL If the judging result of the step S24 is no, the method performs step S25 to create a second gap GP2. If the judging result of the step S23 is no, the judging result of the step S24 is yes, or step S25 is done, the method will perform steps S26 and S27 to turn on SB detection and judge whether SB is detected. If the judging result of the step S27 is no, the method will back to the step S23; if the judging result of the step S27 is yes, the method will perform step S28 to decode the synchronization channel (SCH). Then, the method will perform step S29 to judge whether the basestation identity code (BSIC) and the radio frequency neutralizer (RFN) are decoded from the synchronization channel (SCH).

If the judging result of the step S29 is no, the method will back to the step SI 7; if the judging result of the step S29 is yes, the method will perform step S30 to store the BSCI and RFN and step S31 to judge whether CREATE GAP =1. If the judging result of the step S31 is yes, the method performs step S32 to create a third gap GP3; if the judging result of the step S31 is no or the method finishes the step S32, the method will perform step S33 to turn on the broadcast control channel (BCCH) decoder.

Then, the method performs step S34 to judge whether system information 3 (SI3) is decoded. If the judging result of the step S34 is no, the method will back to the step S31 ; if the judging result of the step S34 is yes, the method will perform step 35 to store the Cell lD, LAC (Location Area Code), and PLMN ID found and step S36 to judge whether SI 13 is present. If the judging result of the step S36 is no, the method will perform step S42 to send the GSM sniffer request; if the judging result of the step S36 is yes, the method will perform step S37 to judge whether CREATE GAP =1. If the judging result of the step S37 is yes, the method performs step S38 to create a fourth gap GP4; if the judging result of the step S37 is no or the method finishes the step S38, the method will perform step S39 to turn on the broadcast control channel (BCCH) decoder. Then, the method will perform step S40 to judge whether system information 13 (SI13) is decoded. If the judging result of the step S40 is no, the method will back to the step S37; if the judging result of the step S40 is yes, the method will perform the step S41 to store routing area code (RAC) and the step S42 to judge whether the receiver have finished sniffing all required ARFCN. If no, the method will back to the step SI 2; if yes, the method will perform step S43 to issue GSM SNIFFER CNF (GSM Sniffer Confirmation) message to the caller.

For WCDMA interferer, the minimum operating window can be proposed for in-band sniffing, as shown in FIG. 13. GPl must be greater than 10ms. GP2 consists of a series of gaps each is 256 chips wide and as an integer multiple of 15. GP3 must be greater than 10ms. GP4 must be greater than 80ms (typical for MIB) and can be up to 320ms (typical for SIB3) or more. Actual gap sizes are for further study. The key is: these gaps should not interfere with normal femtocell operation during an intra-frequency stealth mode.

FIG. 14A-14C show the flowchart of an example of complete WCDMA continuous sniffing operation method. As shown in FIG. 14A-14C, in step S50, the method receives the WCDMA SNIFFER REQ (WCDMA sniffer request) message from the caller. In step

551, the receiver starts reading the U ARFCN (UTRA Absolute Radio Frequency Channel Number) and NCL (Neighboring Cell List) that it is supposed to scan if available. In step

552, the method determines whether the receiver is already powered on or not, it selects a "power on mode" or "normal mode"; the receiver will decide whether there is a need to create transmission gap. In step S55, for "power on" mode, the method sets CREATE GAP = 0. In step S53, for "normal mode", the method selects either "Cell Search Only mode" or "Complete Cycle mode". In step S54, for "Cell Search Only mode", the receiver selects either In-band mode or Out-of-band mode. If in-band mode, the method will perform step S59 to set CREATE GAP = 1. If out-of-band mode, the method will perform step S58 to determine whether the second RF is available for scanning. If the second RF is available for scanning, the method will perform step S55 to set CREATE_GAP = 0; if the second RF is not available for scanning, the method will perform step S59 to set CREATE GAP = 1. In step S56, for "Complete Cycle mode", the receiver sends "CELL RESET _REQ" message to the higher layer request a system reset. In step S57, the method will receive the "CELL RESET CONF" message from the higher layer. Then, the method performs the step S55, the receiver set CREATE GAP = 0.

In step S60, the method judges whether CREATE GAP = 1. If yes, the method performs step S61 to create the first gap GP1. If no or the step S61 is done, the method performs step S62 to turn on primary synchronization channel (PSYNC) searcher. Then, the method performs step S63 to determine whether the slot boundary is detected. If the slot boundary is detected, the method performs step S64 to obtain the UARFCN Band Indicator of the FB detected, and store it for all 32 candidates in the candidate set and step S65 to pick a candidate from the candidate set. If the slot boundary is not detected, the method will back to the step S60.

In step S66, the method determines whether CREATE_GAP = 1. If yes, the method will perform step S67 to create the second gap GP2. If no or the step S67 is done, the method will perform step S68 to turn on Secondary Synchronization Channel (SSYNC) search. If the frame boundary is not detected, the method will back to the step S66. If the frame boundary is detected, the method will perform step S70 to store the code group number found.

In step S71, the method determines whether CREATE GAP = 1. If yes, the method will perform step S72 to create the third gap GP3. If no or the step S72 is done, the method will perform the step S73 to judge whether the Neighboring Cell List (NCL) is available. If the Neighboring Cell List (NCL) is not available, the method will perform step S74 to turn on the scrambling code decoder. If yes or the step S74 is done, the method will perform step S75 to turn on parameter estimations to get PSC (Primary Scrambling Code), RSSI (Received Signal Strength Indicator), RSCP (Received Signal Code Power), CPICH (Common Pilot Channel) Ec/Ior (Energy per chip to Interference ratio) and store them.

In step S77, the method determines whether it is in "Complete Cycle" mode. If it is in "Complete Cycle" mode, the method will perform step S78 to determine whether CREATE GAP = 1. If CREATE GAP = 1, the method will perform step S80 to create the fourth gap GP4. If CREATE GAP = 0 or step S80 is done, the method will perform step S79 to turn on PCCPCH (Primary Physical Common Control Channel) receiver and BCH (Broadcast Channel) decoder.

In step S81, the method determines whether the MIB (Master Information block), SIB1 (System Information Block), SIB2, or SIB3 are decoded. If yes, the method will perform step S82 to store the RAC, LAC, Cell ID, PLMN ID obtained from the information blocks. If no or step S82 is done, the method will perform step S83 to send message "RAKE_REQ" to higher layer. If the judging result of the step S77 is no or the step S83 is done, the method will perform step S84 to run through all 32 candidates in the candidates set. If no, the method will back to the step S65; if yes, the method will perform step S85 to determine whether all UARFCN scanning is finished. If no, the method will back to the step S52. If yes, the method will perform step S86 to send "WCDMA SNIFFER CNF" message to higher layer, and then end the procedure.

Compared to the prior arts, the invention can address the above-mentioned random problem by adding some smartness into the femtocell/small cell devices to narrow the transmit power pattern and hence reduce unwanted radiation and its effect to the existing network elements and address the above-mentioned random on/off nature by adjusting certain transmit characteristic of these small basestations (in the following figure, we call them basestation X) in the goal to optimize some chosen network performance metrics. In addition, the invention can also alleviate the above-mentioned uplink interference by designing the basestation to be aware of such interference and actively reject it or passively avoid it and improve the robustness of the TFCI decoder by reducing its block error rate (BLER).

With the example and explanations above, the features and spirits of the invention will be hopefully well described. Those skilled in the art will readily observe that numerous modifications and alterations of the device may be made while retaining the teaching of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.

Claims

What is claimed is:

1. An interference suppression apparatus for a femtocell, the femtocell having a minimum of two received antenna and RF paths and a minimum of two transmit antenna and RF paths, the interference suppression apparatus comprising:

a monitoring device, for monitoring interfering signals from different directions, the monitoring device comprising:

a forming module, for forming a set of complex number pairs;

a selecting module, coupled to the forming module, for selecting a first complex number pair from the set of complex number pairs; and

a RSSI calculating module, coupled to the selecting module, for calculating a first

RSSI value corresponding to the first complex number pair;

wherein then the selecting module selects a second complex number pair next to the first complex number pair from the set of complex number pairs, and the RSSI calculating module calculates a second RSSI value corresponding to the second complex number pair.

2. The interference suppression apparatus of claim 1, wherein if the set of complex number pairs comprises N complex number pairs, the RSSI calculating module will calculate N RSSI values corresponding to the N complex number pairs in order, and then the monitoring device will select a specific complex number pair corresponding to a minimum RSSI value from the N complex number pairs.

3. The interference suppression apparatus of claim 2, wherein the monitoring device generates a weight pair according to the specific complex number pair, and the weight pair is adjusted gradually from one coordinate to another coordinate in a Markov chain.

4. An interference suppression method for a femtocell to monitor interfering signals from different directions, the femtocell having a minimum of two received antenna and RF paths and a minimum of two transmit antenna and RF paths, the interference suppression method comprising steps of:

forming a set of complex number pairs;

selecting a first complex number pair from the set of complex number pairs;

calculating a first RSSI value corresponding to the first complex number pair;

selecting a second complex number pair next to the first complex number pair from the set of complex number pairs; and

calculating a second RSSI value corresponding to the second complex number pair.

5. The interference suppression method of claim 4, wherein if the set of complex number pairs comprises N complex number pairs, the interference suppression method will calculate N RSSI values corresponding to the N complex number pairs in order, and then select a specific complex number pair corresponding to a minimum RSSI value from the N complex number pairs.

6. The interference suppression method of claim 5, wherein the interference suppression method generates a weight pair according to the specific complex number pair, and the weight pair is adjusted gradually from one coordinate to another coordinate in a Markov chain.

7. An interference suppression apparatus for a femtocell, the femtocell can be arbitrarily powered on or off, the interference suppression apparatus comprising:

a gateway/aggregator, coupled to the femtocell;

a configuration server, coupled to the gateway/aggregator, for using a protocol to communicate with the femtocell; and

an optimization module, coupled to the configuration server, for receiving an information related to the femtocell from the configuration server and continuously adjusting parameters of the femtocell according to the information related to the femtocell to optimize system performance metrics.

8. The interference suppression apparatus of claim 7, wherein the information related to the femtocell is selected from a neighboring basestation ID of the femtocell, a neighboring basestation RSSI value of the femtocell, a quality indicator of a mobile station connected to the femtocell, a power control command of the femtocell for a mobile station connected to the femtocell, and a neighboring basestation current antenna angles and azimuth a neighboring basestation.

9. The interference suppression apparatus of claim 7, wherein the optimization module decides antenna phase multiplier, azimuth multiplier, angle multiplier, and transmission power.

10. The interference suppression apparatus of claim 7, wherein the system performance metrics optimized by the optimization module comprise system throughput, perceived interference, user quality indicator, and RSSI value reported by mobile station.

11. An interference suppression method for a femtocell, the femtocell can be arbitrarily powered on or off, the interference suppression method comprising steps of: using a protocol to communicate with the femtocell;

receiving an information related to the femtocell receiving an information related to the femtocell from the configuration server; and

continuously adjusting parameters of the femtocell according to the information related to the femtocell to optimize system performance metrics.

12. The interference suppression method of claim 11, wherein the information related to the femtocell is selected from a neighboring basestation ID of the femtocell, a neighboring basestation RSSI value of the femtocell, a quality indicator of a mobile station connected to the femtocell, a power control command of the femtocell for a mobile station connected to the femtocell, and a neighboring basestation current antenna angles and azimuth a neighboring basestation.

13. The interference suppression method of claim 11, wherein the system performance metrics optimized by the optimization module comprise system throughput, perceived interference, user quality indicator, and RSSI value reported by mobile station.

14. An interference suppression apparatus for a femtocell, the interference suppression apparatus comprising:

at least one receiving antenna;

at least one RF receiver, coupled to the at least one receiving antenna;

at least one multi-channel channel estimation module, coupled to the at least one RF receiver, for receiving a signal from the at least one RF receiver and forming an augmented channel matrix;

at least one fast Fourier transform (FFT) module, coupled to the at least one multichannel channel estimation module, for receiving output of the at least one multi-channel channel estimation module to form a transform domain signal;

a correlation matrix calculator, coupled to the at least one FFT module, for generating a calculated correlation;

a correlation matrix estimator, coupled to the correlation matrix calculator and the at least one RF receiver, for receiving output of the at least one RF. receiver to estimate a received signal correlation and converting the received signal correlation to transform domain, the correlation matrix estimator receiving the calculated correlation from the correlation matrix calculator and subtracting the calculated correlation from an estimated correlation matrix; and

a joint detection module, coupled to the at least one RF receiver, the at least one FFT module, the correlation matrix calculator, and the correlation matrix estimator, for jointly detecting user signals.

15. The interference suppression apparatus of claim 14, wherein the joint detection module forms a first matrix according to outputs of the correlation matrix calculator and the correlation matrix estimator and processes the first matrix and outputs of the at least one multi-channel channel estimation module and the at least one RF receiver to jointly detecting the user signals.

16. The interference suppression apparatus of claim 15, wherein the output of the at least one RF receiver is multiply by complex conjugate of the at least one multi- channel channel estimation module and then divided by the first matrix to form a first signal and the first signal will be transmitted to an inverse FFT (IFFT) module.

17. A transport format combination indicator (TFCI) decoder used in universal mobile terrestrial system (UMTS), the TFCI decoder comprising:

a mask generator, for generating masks;

a TFCI decoder engine, coupled to the mask generator and a symbol demodulator, for receiving the masks and an input from the symbol demodulator to generate a vector of outputs;

a decision device, coupled to the TFCI decoder engine, for searching a maximum value of the outputs and recording the maximum value, index points to the position of the maximum value in the vector, and a mask ID;

a TFCI value calculator, coupled to the decision device, for receiving the maximum value and reconstructing a TFCI signal;

a TFCI validation module, coupled to the TFCI value calculator, for validate whether the reconstructed TFCI signal has a legal value, if yes, the TFCI validation module setting a valid fail flag to false; and

a transmission time interval (TTI) level TFCI decoder, coupled to the TFCI validation module, for collecting TFCI values from the TFCI validation module belong to the same transmission time interval and outputting a specific TFCI value which gets the highest vote counts among the TFCI values.

18. An interference suppression method for a femtocell to continuously monitor in-band interference to optimize a transmit pattern of the femtocell, the interference suppression method comprising steps of: shutting down a RF transmitter of the femtocell when a measurement is taken to avoid a RF front-end saturation;

if the interference suppression method is operated in a reset mode, normal femtocell transmission and reception being interrupted, a PHY chip itself being reset, all counters and internal states are wiped clean to initiate sniffing, software protocol being not interrupted, after pre-defined sniffing period has passed, the normal femtocell operation will be restarted; and

if the interference suppression method is operated in a stealth mode, normal femtocell operation being muted, all downlink and uplink transmissions/receptions being blanked out, the interference suppression method creating a transmission gap affecting the downlink transmission.

19. The interference suppression method of claim 18, wherein if there is a user in a cell dedicated channel mode, the transmission gap is not greater than 10 radio frames (100 ms).

20. The interference suppression method of claim 18, wherein if there is no user in a cell dedicated channel mode or a cell forward access channel mode, the transmission gap can be greater than 10 radio frames (100 ms).

21. The interference suppression method of claim 18, wherein a user in a cell paging channel mode is ok as the transmission gap can be created in paging channel signals.