While use of power-efficient signaling schemes appears to be effective at compensating for the inherent high path-loss associated with pure diffuse infrared links, it begins to lose its effectiveness as the data rate is increased. At very high data rates, intersymbol interference (ISI) can result in a very high and sometimes irreducible power penalty, preventing the system from operating at a low bit-error probability. We use a link design employing a multibeam transmitter in conjunction with a narrow field-of-view (FOV) direction diversity receiver. The design goal is to eliminate the effect of ISI so that power-efficient signaling schemes such as pulse-position modulation (PPM) can be employed at very high data rates. We also use high-rate Reed-Solomon codes to further increase the power efficiency of PPM signals. The proposed system can be made rate-adaptive through varying modulation level L and/or code rate R without increasing the complexity significantly. This provides a dynamic range large enough to allow efficient utilization of available bandwidth, i.e., to allow portable terminals to communicate at their highest permitted data rate, without sacrificing the quality of service. It is shown that a bit error rate not exceeding 10/sup -9/ can be achieved within the link coverage area with 99% probability at bit rates up to a few hundreds of megabits per second, using transmitted power levels well below 1 W.

For pt.I see ibid., vol.50, no.7, p.1098-1109 (2002). We examine an infrared link composed of a multibeam transmitter and a direction-diversity receiver, employing code combining. The latter represents an added dimension to the conventional diversity concepts, which are limited to combining the individual received symbols. Rate-compatible punctured convolutional codes are used to encode intensity-modulated on-off keying (OOK) optical power, to create an adaptive environment for efficient utilization of channel spectral bandwidth, to provide a means for accurate channel estimation, and to maintain a guaranteed bit error rate (BER) performance at all receiver positions. It is shown that a BER not exceeding 10/sup -9/ with 99% probability can be achieved at bit rates up to a few hundreds of megabits per second, at very low transmitted power levels.

While use of a power efficient signaling scheme appears to be effective in compensating for inherent high path-loss associated with pure diffuse infrared links, it begins to lose its effectiveness as data rate is increased. At very high data rates, intersymbol interference (ISI) can result in a very high and sometimes irreducible power penalty, preventing the system from operating at a low bit-error probability. In this paper, we investigate a link design employing a multi-beam transmitter in conjunction with a narrow field-of-view (FOV) direction diversity receiver. The design goal is to eliminate the effect of ISI, so that power efficient signaling schemes such as pulse-position modulation (PPM) can be employed at very high data rates. It is shown that an average bit-error rate (BER) not exceeding 10/sup -9/ can be achieved within the link coverage area with more than 99% probability at bit rates up to a few hundreds of megabits per second, using transmitted power levels well below one Watt.

This paper presents a solution to the problem of providing quality-of-service (QoS) guarantees in an ATM network with wireless links. The suggested resolution to this problem is through selectable-rate channel coding. In particular, rate-compatible punctured convolutional (RCPC) codes are employed due to their several attractive features that make them favorable for ATM-based applications. In this adaptive scheme, the RCPC code rate is intelligently varied in response to the channel's nonstationary behavior according to the channel state information (CSI). This yields an equivalent stationary channel with a fixed and predetermined bit-error rate (BER) in which QoS can be guaranteed. In order to demonstrate the performance of RCPC codes in providing and maintaining the QoS, a lognormal fading channel model is assumed. An architecture is proposed for the wireless link. This system assumes that accurate CSI is available at the transmitter as well as at the receiver. Bandwidth utilization of the proposed system under various channel conditions is then evaluated, plotted, and discussed.