Ari Hämäläinen

WCDMA common pilot power control for load and coverage balancing

The paper validates the feasibility of automating the setting of common pilot power in a WCDMA radio network. The pilot automation improves operability of the network and it is implemented with a control software aiming for load and coverage balancing. The control applies measurements of base station total transmission power of neighboring cells and terminal reports of received pilot signal level to determine the pilot qualification. The pilot power of a cell is periodically updated with simple heuristic rules in order to improve the load and coverage balance. The approach was validated using a dynamic WCDMA system simulator with a deployment of macro and micro cells on a city region whose measured propagation characteristics were incorporated into the model. The results showed that the proposed control method balanced load and coverage and improved the air interface performance measured as a function of packet throughput.

Verification of WCDMA radio network planning prediction methods with fully dynamic network simulator

The WCDMA based cellular systems offer variability of packet and circuit switched services and therefore are more complicated to plan and control than todays networks. For an operator, it is essential to utilise all possible resources to improve the capacity and quality of service of the radio network. To guarantee the optimum performance, the operator must have means to visualise the offered services and their quality in the radio network-planning phase. The proposed static prediction method is compared with dynamic analysis results. Based on this comparison it is demonstrated that the static prediction method is accurate enough for the radio network planning process. Example results of active set size, dominance and transmit power distributions from both simulators are presented.

Convolutional decoding using recurrent neural networks

We show how recurrent neutral network (RNN) convolutional decoders can be derived. As an example, we derive the RNN decoder for 1/2 rate code with constraint length 3. The derived RNN decoder is tested in Gaussian channel and the results are compared to results of optimal Viterbi decoder. Some simulation results for other constraint length codes are also given. The RNN decoder is tested also with the punctured code. It is seen that RNN decoder can achieve the performance of the Viterbi decoder. The complexity of the RNN decoder seems to increase only polynomially, while in Viterbi algorithm the increase is exponential. Also, the hardware implementation of the proposed RNN decoder is feasible.

Novel use of channel information in a neural convolutional decoder

A neural convolutional decoder which exploits the channel information is introduced. The method uses a recurrent neural network, tailored to the used convolutional code and the channel model. No supervision-besides possible channel estimation-is required. Also, no distinct equalizer is needed. As an example, we show the structure of the neural decoder for 1/2 rate code with constraint length 3 in a two-path channel environment. For testing, the 1/2 rate code with constraint length 5 is used in two-path fading channels. The simulation results show that the proposed decoder works well compared to the traditional way of using some equalizer and the Viterbi decoder. The hardware implementation of the neural decoder seems feasible and its complexity increases only polynomially while in Viterbi algorithm the complexity increases exponentially as a function of the constraint length.

WCDMA adjacent channel interference requirements

In this paper the effects of adjacent channel interference on WCDMA capacity are studied for two adjacent frequency operators as well as for a hierarchical cell structure to be used by one operator at adjacent frequency channels. The effect of adjacent channel interference on the system capacity is studied with practical loading of the WCDMA system. Also a new model for adjacent channel interference in the uplink is introduced taking into account the behaviour of a practical WCDMA terminal power amplifier. A goal of this paper is to find out the adjacent channel interference requirements for the WCDMA system.

Auto-tuning of service-specific requirement of received EbNo in WCDMA

The paper validates the feasibility of auto-tuning of service- and bit rate-specific EbNo (E/sub b//N/sub 0/) requirements. The EbNo requirement is the level of the received bit energy to the interference and noise density that the receiver equipment requires for proper decoding of the signal. The planned EbNo values can be used e.g., to scale the powers when the service is varying. The proposed methods, one for uplink and one for downlink, are tested using a dynamic WCDMA system simulator with a deployment of macrocells on a city region whose measured propagation characteristics have been incorporated into the model. The results show that the proposed methods tune the initially incorrect planned EbNo values so that the throughput in the system increases.

Adaptive power increase estimation in WCDMA

The paper suggests a method of predicting changes in the cell powers of a WCDMA network due to resource allocations such as admitting new users or changing bit rates. The method facilitates load control, admission control, and packet scheduling. The method makes no strict assumptions about the function that maps load influential parameters and changes in them to changes in the total received interference or the total transmission power of the cell. Instead, the method learns the mapping by monitoring power changes that are responses to performed resource allocations. The estimation of the unknown function is implemented with the kernel regression. The method was validated using a dynamic WCDMA system simulator with a deployment of micro cells on a city region whose measured propagation characteristics were incorporated into the model. The results showed that the proposed adaptive power increase estimation (PIE) method improved packet transport performance in comparison to a previously published fixed-formula PIE.

Performance of two neural receiver structures in the presence of co-channel interference

Real communication channels with multipath propagation, interference and possible nonlinearities pose a difficult problem to the detecting receiver. This paper deals with neural approaches to solve those difficulties. Two types of neural networks, self-organizing map and radial basis functions have been studied. The results show that, while there are no actual benefits in using neural receivers in simple white noise Gaussian channels, the performance in nonlinear channels is much better with these new approaches than with the traditional ones.

A ranging system for an IEEE 802.11 OFDM transceiver

In this paper we propose a method for distance estimation between two transceivers. The method is based on controlling transmission times of packets and calculation of the relative phase-shifts of the OFDM subcarriers. We demonstrate the system using simulations of a bit and cycle accurate VHDL model of an IEEE 802.11a WLAN transceiver and the distance measurement algorithm.

Complexity Reduction in Probabilistic Neural Networks

Probability density estimation using the probabilistic neural network or the kernel method is considered. In its basic form this method can be computationally prohibitive, as all training data need to be stored and each individual training vector gives rise to a new term of the estimate. Given an original training sample of size N in a d-dimensional space, a simple binned kernel estimate with O(Nd/d+4) terms can be shown to attain an estimation accuracy only marginally inferior to the standard kernel method. This can be taken to indicate the order of complexity reduction generally achievable when a radial basis function style expansion is used in place of the probabilistic neural network.