The study compares two exercise strategies in people with knee osteoarthritis. The first is aerobic exercise on a recumbent cycle ergometer, with effort controlled by the Borg scale. The second is a therapeutic exercise program with resistance, neuromuscular, mobility, and balance components. The primary objective is to determine which approach produces greater exercise induced hypoalgesia, measured by the increase in pressure pain threshold at the knee within the session across four anchor sessions during a ten week program. This is a randomized clinical trial with two parallel arms. Ninety participants, between forty five and seventy five years of age, will be allocated to one of the two groups. Allocation sequence concealment will be ensured using opaque envelopes. The outcome assessor will remain blinded to group allocation. The interventions will take place three times per week for ten weeks. Assessments will be performed at baseline and after ten weeks. Within session measurements will also be taken in weeks one, four, seven, and ten. The primary outcome is the intra session change in pressure pain threshold at the medial compartment of the knee. Secondary outcomes include exercise induced hypoalgesia in the quadriceps and trapezius, chronic adaptation of pressure pain threshold, conditioned pain modulation, pain intensity, pain self efficacy, knee related health status, functional performance, quadriceps strength, global perceived effect, and adherence. The planned sample size is forty five participants per group. The analysis will follow the intention to treat principle. The primary outcome will be compared using linear mixed models with group and time effects. Clinical outcomes at T1 will be adjusted for baseline values. Estimates, confidence intervals, and interpretation in light of minimally important differences will be reported.