In this paper, measurement results are presented providing experimental support for the validity of a theoretical framework for analysis of heterogeneous systems of interfering packet radio networks. Specifically, an analytical Bluetooth interference model is considered in this paper, where calculated distributions of energy quantities from interfering Bluetooth networks are compared with distributions obtained from measurements. The measurements indicate that the closed form expressions for the distributions of the received interfering energy in the analytical model capture important mechanisms in real Bluetooth networks. It is also evident from the measurements that adjacent channel interference must be taken into account when analyzing interfering Bluetooth networks

While the use of radio technology for wireless data communications has increased rapidly, the wide variety of radio interfaces being used has made interference investigations hard to perform. With that in mind, we present a novel approach for analyzing packet radio communications, applicable to interfering heterogeneous networks, which leads to tractable analytical expressions. The core of the approach is an analytical framework modeling each network with individual properties for the packet types and the channel sets used, while taking path loss between nodes into account. Furthermore, we present a derivation of closed-form expressions for the throughput of the networks, thus allowing for the investigation of important mechanisms limiting network and system performance; the expressions enable fast and flexible analysis to be performed without extensive computer simulations or measurement campaigns. To illustrate the use of the framework and the strength of the closed-form expressions, we analyze a heterogeneous example system consisting of the IEEE 802.llb network and multiple Bluetooth networks that use multiple packet types. In the analysis, we also take the adjacent channel interference into account when calculating network throughput as functions of number of interferers in the system.

While the use of radio technology for wireless data communications has increased rapidly, the wide variety of radio interfaces being used has made interference investigations hard to perform. With that in mind, we present a novel approach for analyzing packet radio communications, applicable to interfering heterogeneous networks, which leads to tractable analytical expressions. The core of the approach is an analytical framework modeling each network with individual properties for the packet types and the channel sets used, while taking path loss between all network nodes into account. Furthermore, we present a derivation of closed-form expressions for the throughput of the networks, thus allowing for the investigation of important mechanisms limiting network and system performance. The expressions enable fast and flexible analysis to be performed without extensive computer simulations or measurement campaigns. To illustrate the use of the framework and the strength of the closed-form expressions, we analyze a heterogeneous example system consisting of one IEEE 802.11b network and multiple Bluetooth networks that use multiple packet types. In the analysis, we also take the adjacent channel interference into account when calculating network throughput as functions of the number of interferers in the system

The increasing use of wireless technology utilizing unlicensed frequency bands calls for more in-depth analysis of interference and coexistence between systems. In this paper a framework is presented for detailed analysis of the performance of coexisting networks in shared frequency bands. The framework allows for multiple packet lengths to be used by the communicating devices and the analysis is performed with respect to the received interfering energy, which in effect leads to a link budget analysis on a packet basis. A system of interfering Bluetooth piconets is analyzed to illustrate the use of the framework and the conceptual difference in basing the analysis on link budgets, rather than on packet collisions. Furthermore, some indications on throughput saturation in the analyzed Bluetooth system are presented.

The increasing use of unlicensed frequency bands for radio communications calls for investigations of the interference between coexisting radio networks. The wide variety of radio interfaces makes such investigations hard to perform. We present a general framework for the analysis of heterogeneous interfering radio networks, where the networks are modeled with individual properties for the packet types used, transmitted power distributions in time and frequency from the packet transmissions, and with path loss between network nodes. By using closed form expressions for the throughput of the networks, important mechanisms limiting their performance can be investigated. The closed form expressions enable fast and flexible analysis to be performed without extensive computer simulations. To illustrate the use of the framework we analyze an example system of interfering IEEE 802.11b and Bluetooth networks.

With the increasing use of short-range wireless devices for high-data-rate communication in the shared frequency bands, the level of interference can be expected to increase. The ability to estimate the performance degradation of these devices due to increasing interference is, therefore, important. In this paper, the throughput of devices that perform frequency hops after each transmitted packet in order to achieve diversity is investigated. The system model allows for an analysis of systems where packets of variable durations are used, and the throughput derivation is based on the assumption that collisions result in a total loss of the data in the colliding packets. The resulting expression for the throughput is given as a function of the number of frequency channels used for frequency hopping, the number of interfering networks, the durations of the packet types available, and the probability of networks selecting a certain packet type for transmission. An approximation of the exact expression for the throughput is also derived, and the results are applied to an example system consisting of Bluetooth piconets.

The impact of the quantization of SNR measurements on the throughput of a multiuser diversity scheme for constant-rate transmission is investigated under a block-Rayleigh fading assumption. In the downlink, each user measures its SNR, quantizes it, and feeds it back to the transmitter, which transmits a packet to the user with the highest quantized SNR. In the case of several users having the same quantized SNR, one of them is selected at random. It is concluded that using only a few quantization levels can yield a throughput that is only slightly less than the throughput obtained by using unquantized feedback.

The impact of interference from Bluetooth networks on the throughput of IEEE 802.11 FHSS networks is investigated. This is done by deriving an analytical approximation of the throughput of slow frequency-hopping systems. The derivation in itself provides valuable insights into the mechanisms of interference between systems employing the frequency-hopping technique. In deriving the approximation, it is assumed that packet collisions result in total loss of all information contained in the packets involved in the collisions, regardless of the distance between the networks. The results indicate that the Bluetooth networks may have a negative effect on the throughput of an IEEE 802.11 network using long packet types.

This paper describes a strategic research program in personal computing and communication. The program was initiated by the Swedish Foundation for Strategic Research and started in 1997. The main objectives for the program are to promote multidisciplinary research in computing and communication and to initiate cooperation between research groups within academia as well as between industry and academy. The main activities in the program are multidisciplinary research projects covering several universities and industries, and a distributed graduate school comprising 30 Ph.D. students today and a planned increase to more than 50 students by the end of 1999.

This paper describes the implementation of a switch-level model in a waveform-relaxation-based concurrent circuit simulator. The model requires much less time for model evaluation compared to continuous models, while still giving sufficient accuracy. Multicomputer runs have shown speedups in the range from 5 to 10, corresponding to efficiencies between 15 and 30%; more than 100 computing nodes have been efficiently used, with less than 4 circuit nodes per computing node.>