Stephen Craig

Adaptive thresholds for AMR codec mode selection

The speech codecs from the adaptive multi-rate (AMR) codec family enable provisioning of excellent speech quality, at the same time providing a way forward towards state-of-the-art, spectrally efficient, high capacity cellular networks. One straightforward way to characterize the benefit of AMR speech codecs is that the robustness to interference and noise in radio networks is increased and that this advantage over other, nonadaptive, speech codecs can be capitalized on in several different ways, e.g., by enhancing speech quality or improving system capacity. In this paper, improved mode adaptation, where codec mode switching thresholds are adaptive to radio conditions, is discussed. Example simulations show that an adaptive thresholds algorithm applied to GSM can significantly improve objective speech quality. Corresponding improvements were also found in informal listening tests.

Single antenna interference rejection in GSM/EDGE networks

While previous interference rejection techniques suitable for the widespread GSM/EDGE standard either require multiple antennas or are highly complex, this paper investigates the performance of a low-complexity single antenna interference rejection technique suitable for implementation in todays GSM terminals. The study is performed both on the link level in different scenarios and on the system level with fully dynamic radio network simulations. The link level simulations show significant gains. For example, with a single co-channel interferer, the gain over a conventional receiver is up to 10 dB. On the system level it is shown that the introduction of the proposed interference rejection method in all terminals gives a large increase in downlink speech capacity in a tightly planned GSM network. The strong but sporadic interference in such a network is an ideal environment for efficient interference rejection.

Collision properties of GSM hopping sequences

Properties of GSM pseudo-random frequency hopping sequences (FHSs) are investigated and it is concluded that they differ significantly from the properties of ideal random sequences. Such discrepancies are evident from correlations as well as collision rates between sequences. It is shown that the collision rate between two FHSs depends heavily on certain FHS parameters such as the hopping sequence number (HSN) and the mobile allocation index offset (MAIO). Another important parameter with significant impact on the collision rate is the frame number (FN) difference between two cells. A consequence is that co-channel interference between cells with fractional loading can be reduced by careful HSN/FN/MAIO planning based solely on collision rate information. This is demonstrated in examples involving two cells.

Uplink Load Based Scheduling for CDMA Systems

Cellular third-generation wideband code division multiple access (WCDMA) systems must meet interference constraints to maintain coverage when uplink traffic is scheduled. The interference constraints can be expressed in the form of load factor constraints but current scheduling algorithms are typically not formulated in terms of these variables. The current paper shows how a cost function for the scheduling problem can be expressed directly in terms of load factors. A near optimal solution is developed for the resulting problem.

Coverage Improvements for Enhanced Uplink

In UMTS Enhanced Uplink (EUL), 2 and 10 ms TTIs are supported, which substantially decreases latency and increases data rates compared to DCH. However, due to the shorter TTI compared to DCH, coverage may be an issue in situations where the use of multiple re-transmissions is not feasible. This paper evaluates four possible methods to increase the EUL coverage: Configurable minimum data power offset, Freezing the Outer Loop Power Control (Freezing OLPC), Autonomous Retransmission (AR) and Improved Layer 2 (Improved L2). The coverage gains with all these methods are considerable compared to EUL Release 6 (R6). By tuning the minimum data power offset for VoIP, it is found that a coverage gain of 4-5 dB is achievable. With Improved L2 and 120 bits as the smallest transport block size, a coverage gain of up to 7 dB is possible.

High capacity strategies for GSM/EDGE - impacts on data traffic performance

GSM is the worlds most widespread cellular technology. With EDGE packet data, an attractive solution to offer 3rd generation cellular services in existing GSM networks is provided. The introduction of new services into an existing network typically leads to increased capacity demands, since more users of different service types need to coexist in a limited spectrum, while the existing speech service is required to function at least as well as before. In GSM/EDGE systems, the highest capacity is reached in interference-limited scenarios, typically realized with tight frequency reuse and fractional loading; see for example P. de Bruin et al. (IEEE Vehicular Society News, vol.49, no.1, 2002). In such systems, methods for mitigating interference are needed to achieve maximum performance. Strategies for frequency hopping, inter-cell synchronization and MAIO (mobile allocation index offset) planning becomes increasingly important. The paper examines the impact these methods have on EDGE packet data performance. It is shown that packet data traffic responds differently than speech to interference diversity, mainly due to differences in channel coding and radio protocol characteristics. However, it is demonstrated that the intercell synchronization and MAIO planning strategies preferred for pure speech networks also offer substantial performance improvements for EDGE packet data.

Adaptive antenna systems and EGPRS data protocol aspects

Antenna systems, which can adapt their radiation pattern dynamically, are often referred to as adaptive antenna systems. By using adaptive antenna systems in a cellular network the interference can be minimized since it is possible to transmit only in the direction of the mobile that needs the information. However, a complication in some packet data networks is that the data packets transmitted on the downlink may contain information intended for two or more mobiles, which might be located in different parts of the cell. An example of this is the data payload and the signaling message for uplink transmission permission in GPRS/EGPRS systems. A number of design concepts that mitigate this transmission direction issue are presented and evaluated in a cellular network simulator. The most promising concept preserves the large system gains from adaptive antenna systems, offering performance within 5 percent of that of an ideal adaptive antenna system without any transmission direction complications.