This paper proposes a distributed quality-of-service (QoS) routing scheme for ad hoc networks. In ad hoc networks, communication occurs wirelessly without wired base stations, and nodes communicate through multiple hops. QoS routing is challenging due to the dynamic and imprecise nature of the network. The proposed scheme selects a path with sufficient resources to meet QoS requirements, such as delay or bandwidth, while considering cost optimality. The algorithms work with imprecise state information and use parallel path searching to find the most qualified path. Fault-tolerance techniques are used to maintain routing paths when nodes move, join, or leave the network. The algorithms consider both QoS requirements and cost optimality to improve overall network performance. Simulations show that the scheme achieves high call-admission ratio and low-cost paths with modest routing overhead. The algorithms can tolerate a high degree of information imprecision. The paper discusses the system model of ad hoc networks, including stationary and transient links, and QoS state metrics. It also describes the routing problems of delay-constrained and bandwidth-constrained routing. The paper introduces a ticket-based probing approach for distributed QoS routing, which uses tickets to search for feasible paths. The number of tickets is determined based on the QoS requirements, and the distribution of tickets is based on delay or cost. The algorithm uses multiple paths to increase the probability of finding a feasible path and considers both QoS and cost optimality. The paper also discusses rerouting, soft states, and local multicast techniques for maintaining and repairing routing paths in dynamic environments. The proposed scheme is efficient and effective in ad hoc networks with relatively stable topologies.This paper proposes a distributed quality-of-service (QoS) routing scheme for ad hoc networks. In ad hoc networks, communication occurs wirelessly without wired base stations, and nodes communicate through multiple hops. QoS routing is challenging due to the dynamic and imprecise nature of the network. The proposed scheme selects a path with sufficient resources to meet QoS requirements, such as delay or bandwidth, while considering cost optimality. The algorithms work with imprecise state information and use parallel path searching to find the most qualified path. Fault-tolerance techniques are used to maintain routing paths when nodes move, join, or leave the network. The algorithms consider both QoS requirements and cost optimality to improve overall network performance. Simulations show that the scheme achieves high call-admission ratio and low-cost paths with modest routing overhead. The algorithms can tolerate a high degree of information imprecision. The paper discusses the system model of ad hoc networks, including stationary and transient links, and QoS state metrics. It also describes the routing problems of delay-constrained and bandwidth-constrained routing. The paper introduces a ticket-based probing approach for distributed QoS routing, which uses tickets to search for feasible paths. The number of tickets is determined based on the QoS requirements, and the distribution of tickets is based on delay or cost. The algorithm uses multiple paths to increase the probability of finding a feasible path and considers both QoS and cost optimality. The paper also discusses rerouting, soft states, and local multicast techniques for maintaining and repairing routing paths in dynamic environments. The proposed scheme is efficient and effective in ad hoc networks with relatively stable topologies.